Parallax barrier
More actions
A parallax barrier is an optical mask of alternating opaque and transparent strips (or, in two-dimensional versions, an array of pinholes) placed a short distance in front of or behind an image, so that each of the viewer's eyes sees a different set of pixel columns. Because the two eyes look through the slits from slightly different angles, an image made of interleaved left-eye and right-eye strips is separated into a stereo pair, and the viewer perceives depth without wearing glasses.[1][2] It is one of the two main light-directing elements used in autostereoscopic displays; the other is the lenticular sheet of cylindrical lenses.[2][3]
The idea dates to 1896, when Auguste Berthier proposed a barrier to make an autostereogram, and to Frederic E. Ives's 1903 patent for a "parallax stereogram".[2][4] Electronic barriers made of liquid crystal appeared in Sharp's switchable 2D/3D LCDs from 2002 and in the Nintendo 3DS handheld in 2011.[1][5] In virtual reality research, head-tracked barrier displays such as the Electronic Visualization Laboratory's Varrier were built to give a stereoscopic, head-tracked VR view without glasses or worn trackers.[3]
How it works
Sharp Laboratories of Europe describes the barrier as alternating transmissive and non-transmissive columns aligned with the pixel columns of an LCD. The transmissive columns create two regions in space, called viewing windows, in front of the display. Two 2D images are shown on the panel at once, interleaved column by column; when the viewer's right eye is in one window and the left eye in the other, each eye sees only its own image and the pair is fused into a 3D image. The distance at which this works depends mainly on the pixel pitch and on the gap between the barrier and the LCD.[1] In the Dynallax paper from the Electronic Visualization Laboratory, Tom Peterka and colleagues describe its function: the barrier occludes certain regions of the image from each eye while letting other regions through, so that strips of the left-eye image reach only the left eye and strips of the right-eye image only the right eye.[6]
The geometry follows from similar triangles. Ken Perlin, Salvatore Paxia and Joel Kollin give a worked example: if the eyes are a distance E apart (about 2.5 inches) and the display screen is a distance D from the viewer, a light-blocking "picket fence" placed a distance G in front of the screen needs stripes of width EG/D, while the image stripes on the screen behind it are EG/(D - G) wide. With those widths, the left eye sees half of one image through the clear stripes and the right eye half of the other.[7] The barrier period is not exactly a whole multiple of the pixel pitch. In a four-view front-barrier design described by Chen, Qiu and Liu, the barrier pitch is slightly smaller than four times the subpixel pitch so that the light converges into a single view.[2]
Two-view and multiview barriers
The basic barrier sends two views to two eyes. Changing the barrier pitch turns it into a multiview display that sends several views into adjacent zones, so a viewer who moves sideways sees new perspectives (motion parallax) and several people can watch at once. The cost is that both the brightness and the horizontal resolution fall to 1/N, where N is the number of views.[2] The width of the clear slits sets a second trade-off: a wider opening makes the image brighter but lets more light from neighbouring views through, which causes crosstalk. Chen and colleagues give one quarter of the pitch as a common slit width in a four-view design.[2]
Barriers can also be two-dimensional. A patent by Douglas Lanman and colleagues at the Massachusetts Institute of Technology describes conventional barriers made of slits as supporting horizontal-only parallax, while a regular grid of pinholes gives both horizontal and vertical parallax; the light attenuation is more severe for pinholes than for slits.[8]
Front and rear barriers
The barrier can sit between the LCD and the viewer (a front barrier) or between the backlight and the LCD (a rear barrier); Sharp states that both produce the same 3D effect.[1] Chen and colleagues note that placing it behind the panel keeps viewers from noticing the black strips.[2] Sharp found the rear arrangement useful for transflective LCDs that work in both transmissive and reflective modes: with a rear barrier, the brightness in reflective mode is identical to an ordinary 2D display and only the transmitted brightness is reduced, whereas a front barrier reduces both.[1]
Limitations
Perlin and colleagues summarised the static barrier's two notable problems in 2000: the observer must remain in a fixed position, and each eye sees only half the horizontal screen resolution.[7] Further problems are:
| Problem | Cause and effect |
|---|---|
| Low brightness | The opaque strips block most of the light. In EVL's Varrier, the barrier was 77.78 percent opaque and 22.22 percent transparent.[3] The MIT patent states that a binary mask of slits or pinholes "severely attenuates" the light of the 3D image.[8] IEEE Spectrum reported that reaching acceptable brightness means raising the backlight, which drains batteries in portable devices.[9] |
| Reduced resolution | Each view gets only a fraction of the pixel columns. Peterka et al. note that in head-tracked static barrier systems such as Varrier, a duty cycle of about three quarters black leaves each eye channel only a quarter of the horizontal screen resolution.[6] |
| Fixed viewing zone | A static barrier is designed for one distance. IEEE Spectrum reported that consumer barrier screens are tuned to an average eye separation of 65 mm and to a single viewer at a set distance, usually around 30 cm, and that the 3D effect fades if that distance is off by as little as 5 cm.[9] |
| Crosstalk (ghosting) | Light meant for one eye reaches the other, producing double images. Peterka et al. measured a minimum ghost level of 5 to 7 percent in both static and dynamic barrier systems at EVL.[6] |
| Wrong-eye views | If the viewer moves away from the head-on position, the left eye can see the image meant for the right eye.[9] Outside the central field of view, a conventional barrier also projects periodic replicas of the central views.[8] |
| Moiré | Interference between the barrier and the pixel grid. Varrier tilted its barrier 7.82 degrees from vertical, an angle chosen to minimise static and dynamic moiré patterns and colour shift.[3] In dual-stacked LCD barriers, viewing one LCD through another can also cause moiré fringes.[8] |
| Distorted 2D content | Ordinary text viewed through a 3D barrier can be distorted, and brightness can vary as the viewer moves between windows, which is why Sharp made its barrier switchable.[1] |
History
Photographic stereograms
In 1896 Auguste Berthier published "Images stéréoscopiques de grand format" in the French weekly Cosmos, which proposed a parallax barrier for making an autostereogram.[2] Frederic E. Ives described a "novel stereogram" in the Journal of the Franklin Institute in 1902.[10] His US patent 725,567, "Parallax stereogram and process of making same", was filed on 25 September 1902 and granted on 14 April 1903. It combines the two images of a stereo pair into one photograph made of fine lines and covers it with "a transparent-line screen consisting of opaque lines with clear spaces between them", superposed on the photograph "with a definite separation therefrom, calculated by reason of the parallax of vision with two eyes", so that the lines belonging to each stereo image are visible to one eye only.[4] EVL's Varrier paper credits Ives with patenting "the first parallax stereogram" in 1903.[3]
Clarence W. Kanolt extended the idea from two views to many. His US patent 1,260,682, "Photographic method and apparatus", filed on 16 January 1915 and granted on 26 March 1918, uses a screen of alternating light-transmitting and light-excluding strips and moves the camera relative to the subject during exposure, so that each of the viewer's eyes sees the scene from a different camera position.[11] Later patents refer to Kanolt's 1918 method as the parallax panoramagram.[12]
Electronic barriers
Liquid crystal panels allowed the barrier itself to become a display. Japan's public broadcaster NHK patented a 3D display in which a second transmissive LCD, laminated to the image panel, draws the barrier stripes electrically, so their density and position can be changed and the barrier can be turned off for 2D viewing; the patent has a priority date of 28 October 1991 and was granted in the United States on 24 May 1994.[12]
Sharp Laboratories of Europe (SLE) began researching 3D displays in 1992 and built its first single-panel 3D display based on the parallax barrier method in 1994. These early panels were "3D-only"; in 1996 SLE added a "sweet spot indicator" to help users find the correct viewing position, and polarisation optics later allowed 8.4-inch and 13.8-inch panels to be switched mechanically between full-resolution 2D and 3D. The first electronically switchable prototype followed in October 2001. Sharp announced mass production of electrically switchable 2D/3D displays in September 2002, and the first product, the SH251iS mobile phone for NTT DoCoMo with a 2.2-inch screen, launched in November 2002.[1] In 2003 Sharp released the PC-RD3D notebook, whose LCD switches between 3D and planar modes without glasses; Sharp describes it as the first notebook computer with a 3D LCD.[13]
The Nintendo 3DS, launched in Japan in February 2011, uses a parallax barrier screen that shows stereoscopic images to a user centred in front of the device.[5] Its upper screen is a 3.53-inch LCD with a resolution of 800 x 240 pixels, and Nintendo states that "400 pixels are allocated to each eye to enable 3D viewing".[14] IEEE Spectrum reported that the screen was made by Sharp and adds a second layer of liquid crystal that forms thin vertical strips, blocking some of the light and directing the rest alternately to the left and right eyes.[9] Chen and colleagues describe the 3DS barrier as a switchable liquid crystal shutter array placed behind the image panel.[2] Smartphones of the same period used the technique too; GSMArena's 2011 review of the HTC EVO 3D describes its glasses-free effect as coming from "the parallax-barrier technology", an extra screen layer that makes each eye see a different set of pixels, with each eye getting half of the horizontal resolution.[15] The New Nintendo 3DS, launched in Japan in October 2014, added "super-stable 3D": it tracks the user's face with the internal camera and an infrared LED and adjusts its dynamic parallax barrier to match, giving a much wider viewing range than the original model.[5]
Later eye-tracked desktop products generally use lenticular optics instead of barriers. Acer's SpatialLabs, announced on 27 May 2021, puts a liquid crystal lenticular lens on top of a UHD panel and tracks the user's head and eyes with a stereo camera,[16] and Samsung's Odyssey 3D G90XF monitor, launched in April 2025, combines eye tracking with a lenticular lens.[17]
| Year | System | Barrier | Notes |
|---|---|---|---|
| 1896 | Berthier autostereogram proposal | Parallax barrier | Published in Cosmos[2] |
| 1903 | Ives parallax stereogram (US 725,567) | Opaque-line screen at a calculated separation | Two-view photograph[4] |
| 1918 | Kanolt (US 1,260,682) | Alternating light-transmitting and light-excluding strips | Many viewpoints recorded by moving the camera[11] |
| 1991 (priority) | NHK electrically generated barrier (US 5,315,377) | Second LCD panel draws the stripes | Switchable between 2D and 3D[12] |
| 2000 | NYU autostereoscopic display (Perlin et al.) | Pi-cell liquid crystal shutter, three phases | Eye-tracked single viewer[7] |
| 2002 | Sharp SH251iS phone | Electrically switchable LC barrier | First product with Sharp's 2D/3D display[1] |
| 2004-2005 | Sandin et al., Varrier (EVL) | Printed film barrier on 35 tiled LCDs | Head-tracked autostereoscopic VR[3][6] |
| 2007 | Peterka et al., Dynallax (EVL) | Front LCD of a dual-stacked LCD | Dynamic barrier, up to two tracked viewers[6] |
| 2011 | Nintendo 3DS | Switchable LC barrier | 800 x 240 screen, 400 pixels per eye[14][5] |
| 2011 | HTC EVO 3D | Extra screen layer | Half horizontal resolution per eye in 3D[15] |
| 2014 | New Nintendo 3DS | Dynamic barrier steered by face tracking | "Super-stable 3D"[5] |
Switchable and dynamic barriers
A static printed barrier cannot be disabled, so the display cannot be converted to 2D use.[6] Sharp's solution was a patterned retarder barrier: in 2D mode both columns of the retarder transmit equally and image quality matches a standard LCD, while in 3D mode a liquid crystal cell rotates the polarisation so that one column transmits and the other is opaque.[1] Chen and colleagues list several other switchable designs: Samsung and LG published barriers made of a switching LCD inserted between the LCD and its backlight; Sanyo used a polymer-dispersed liquid crystal film that is transparent (barrier active, 3D mode) or diffuse (barrier ineffective, 2D mode) depending on the applied field; and Taiwan's Industrial Technology Research Institute stacked two LCDs with a micro-retarder film to form a localized barrier that shows 2D and 3D content side by side, work that won an R&D 100 award in 2010.[2]
A dynamic barrier goes further by moving or reshaping the slits in real time. Perlin, Paxia and Kollin argued in 2000 that no fixed barrier geometry could serve an observer who moves and turns freely, and built a barrier that continually changes the width and position of its stripes as the viewer moves. Their prototype used a custom pi-cell liquid crystal shutter with an active area of 14 x 12 inches and 20 vertical microstripes per inch, placed in front of a rear-projection screen lit by a Texas Instruments DLP projector. The barrier cycled through three phases at 180 Hz; the design aimed to avoid the factor-of-two loss of horizontal resolution of ordinary barriers, and the rapidly moving stripes could not be perceived individually. A retroreflective infrared camera tracker found both eyes, and a Kalman filter smoothed the eye positions; the first prototype showed monochrome images and handled observers up to about five feet away.[7]
Time multiplexing can also recover resolution. Chen and colleagues describe a barrier whose slits shift by half a pitch every 1/120 second, so both eyes see all pixels at a 120 Hz frame rate; they note that most time-multiplexed displays remain two-view because liquid crystal switches slowly.[2] Hideki Kakeya and colleagues proposed in 2018 a barrier that adapts its time division to ambient light: time-division quadruplexing in the dark, where crosstalk is more visible, and triplexing in bright surroundings, which gives a brighter image at the cost of a narrower crosstalk-free viewing zone.[18]
Applications in VR and AR
Autostereoscopic virtual reality
According to the Varrier authors, the CAVE and earlier VR technologies such as head-mounted displays required users to wear stereo glasses and sometimes tracking gear. The Varrier project at the Electronic Visualization Laboratory of the University of Illinois at Chicago set out to remove both. Its authors define VR as requiring head-tracked first-person perspective, large angles of view, stereoscopic display and real-time interactivity, and note that most commercial autostereoscopic displays of the time did not meet those criteria.[3] EVL presented Varrier in conjunction with IEEE VR 2004.[6]
Varrier combines a physical parallax barrier with a "virtual barrier": a model of the barrier inside the rendered scene, used with the OpenGL depth buffer to interleave left-eye and right-eye perspectives in world coordinates.[3] The 2005 system, described in ACM Transactions on Graphics, had the following design:[3]
| Feature | Value |
|---|---|
| Panels | 35 (5 high x 7 wide) NEC 2080 LCD monitors, each 1600 x 1200 pixels and 20 inches diagonal, arranged on a curve of 60 in (1.52 m) radius |
| Field of view | Approximately 120 to 180 degrees |
| Resolution | 11200 x 6000 pixels gross (67 megapixels); about 2500 x 6000 (15 megapixels) net through the barrier |
| Barrier | Pattern printed on photographic film, pitch 1.123 mm (just over 4 LCD pixels of 0.254 mm), 77.78 percent opaque, tilted 7.82 degrees from vertical |
| Viewing distance | 1.02 to 2.24 m, optimum 1.64 m (optical limits 27 to 100 inches, narrowed by tracker coverage) |
| Tracking | Two cameras with infrared illumination panels and artificial neural networks that recognise and track faces at 120 fps; no markers worn |
| Latency | 81 ms end to end (28 ms tracking, 37 ms communication, 16 ms rendering) |
| Computing | 19-node Linux cluster with Nvidia Quadro FX3000 graphics, applications built on CAVELib |
The paper notes that the barrier is not as bright as a lenticular screen but is easily made by printing the pattern onto film.[3] Users of Varrier pointed out its limited spatial resolution, restrictions on how fast the head could move, and the fact that only one person could see the 3D image at a time. EVL's follow-up, Dynallax (IEEE VR 2007), replaced the film with a dual-stacked LCD in which the front panel renders the barrier and the rear panel the VR scene. Varying the barrier period in real time extended the usable distance range, let users switch to 2D, eliminated the need to register a physical barrier, and supported two tracked viewers with four independent eye channels. The penalty was a display with roughly half the brightness and contrast of a static barrier system.[6] A 2010 follow-up by Robert Kooima, Sandin, DeFanti and others used tiled lenticular displays for multi-viewer VR, and describes Varrier as a single-user system that presents exactly two image channels.[19]
Head-tracked barrier displays such as Varrier and Perlin's NYU display render each eye's image from that eye's tracked position, as in head-coupled perspective on ordinary monitors. Perlin and colleagues describe the goal as letting an unencumbered observer walk up to a simulated object and look at it from any distance and angle while it stays in a consistent spatial position.[7][3]
Augmented reality
The Nintendo 3DS came with the built-in AR Games software and AR cards: the player places a card on a flat surface and points the system's outer cameras at it, and targets, objects and Mii characters appear on the 3D screen within the scene in front of the player.[20][5]
Barriers have also been studied for head-up displays. In the 2018 SID Symposium Digest, Matsumoto, Kusafuka, Hamagishi and Takahashi reported a glasses-free 3D head-up display that uses a parallax barrier designed for a magnified virtual image system, with image processing driven by camera-based eye tracking to keep the 3D image correct inside the HUD's eye box.[21]
Research
Barriers as light field displays
Later research treats a barrier display as a simple light field display. In "Content-Adaptive Parallax Barriers" (SIGGRAPH Asia 2010), Lanman, Matthew Hirsch, Yunhee Kim and Ramesh Raskar of the MIT Media Lab showed that any 4D light field emitted by two stacked masks is the tensor product of two 2D masks, so a conventional barrier (a uniform grid of slits or pinholes on the front layer) only achieves a rank-1 approximation of the target light field. Letting both layers take non-binary values optimised for the content by non-negative matrix factorization, and time-multiplexing several mask pairs, increased brightness and refresh rate compared with conventional barriers. Their prototype stacked two 1680 x 1050, 120 Hz LCD panels 1.5 cm apart and supported a field of view of 11 x 7 degrees for 5 x 3 views.[22][8] Gordon Wetzstein, Lanman, Hirsch and Raskar generalised this in 2012 as tensor displays, a family of compressive light field displays built from any stack of time-multiplexed, light-attenuating layers, optimised with non-negative tensor factorization.[23]
The same optics can correct vision. Fu-Chung Huang, Wetzstein, Brian Barsky and Raskar built a vision-correcting display (ACM Transactions on Graphics, 2014) from a pinhole-array barrier mask mounted a small distance in front of an Apple iPod touch 4 screen; the pinhole array generates multiple angular views so that at least two enter the observer's pupil at once, and 4D prefiltering of the content then produces vision-corrected imagery.[24]
Comparison with lenticular lenses
Barrier and lenticular displays do the same job; the Varrier authors call their functions equivalent.[3] The Varrier paper notes that its barrier is not as bright as a lenticular screen, and Chen and colleagues describe the lenticular sheet as a way to overcome the barrier's loss of brightness.[3][2] A barrier, on the other hand, can be made simply by printing a pattern onto film, as in Varrier, or formed by a liquid crystal layer that is switched or reshaped electronically, as in the dynamic and content-adaptive designs described above.[3][22][6]
See also
References
- ↑ 1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 Adrian Jacobs, Jonathan Mather, Robert Winlow, David Montgomery, Graham Jones, Morgan Willis, Martin Tillin, Lyndon Hill, Marina Khazova, Heather Stevenson, Grant Bourhill (2003-04). "2D/3D Switchable Displays". Sharp Technical Journal, no. 4. Sharp Corporation. https://global.sharp/corporate/info/rd/tj4/pdf/4.pdf. Retrieved 2026-10-06.
- ↑ 2.00 2.01 2.02 2.03 2.04 2.05 2.06 2.07 2.08 2.09 2.10 2.11 2.12 Fuhao Chen, Chengfeng Qiu, Zhaojun Liu (2022-01-27). "Investigation of Autostereoscopic Displays Based on Various Display Technologies". Nanomaterials, vol. 12, no. 3, art. 429. MDPI. doi:10.3390/nano12030429. https://pmc.ncbi.nlm.nih.gov/articles/PMC8839869/. Retrieved 2026-10-06.
- ↑ 3.00 3.01 3.02 3.03 3.04 3.05 3.06 3.07 3.08 3.09 3.10 3.11 3.12 3.13 Daniel J. Sandin, Todd Margolis, Jinghua Ge, Javier Girado, Tom Peterka, Thomas A. DeFanti (2005-07). "The Varrier Autostereoscopic Virtual Reality Display". ACM Transactions on Graphics, vol. 24, no. 3 (Proceedings of ACM SIGGRAPH 2005), pp. 894-903. doi:10.1145/1073204.1073279. https://doi.org/10.1145/1073204.1073279. Retrieved 2026-10-06.
- ↑ 4.0 4.1 4.2 Frederic E. Ives (1903-04-14). "US725567A - Parallax stereogram and process of making same". Google Patents. United States Patent Office. https://patents.google.com/patent/US725567A/en. Retrieved 2026-10-06.
- ↑ 5.0 5.1 5.2 5.3 5.4 5.5 "NERD Teams Up With Hardware Experts in Japan to Deliver 'Super-Stable 3d' on the New Nintendo 3DS". Nintendo European Research & Development. Nintendo. 2014-10. https://www.nerd.nintendo.com/optics/2014/10/01/Super_stable_3D.html. Retrieved 2026-10-06.
- ↑ 6.0 6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 Tom Peterka, Robert L. Kooima, Javier I. Girado, Jinghua Ge, Daniel J. Sandin, Andrew Johnson, Jason Leigh, Jurgen Schulze, Thomas A. DeFanti (2007). "Dynallax: Solid State Dynamic Parallax Barrier Autostereoscopic VR Display". 2007 IEEE Virtual Reality Conference, pp. 155-162. IEEE. doi:10.1109/VR.2007.352476. https://doi.org/10.1109/VR.2007.352476. Retrieved 2026-10-06.
- ↑ 7.0 7.1 7.2 7.3 7.4 Ken Perlin, Salvatore Paxia, Joel S. Kollin (2000). "An Autostereoscopic Display". Proceedings of the 27th Annual Conference on Computer Graphics and Interactive Techniques (SIGGRAPH 2000), pp. 319-326. ACM. doi:10.1145/344779.344933. https://mrl.cs.nyu.edu/publications/autostereo/autostereo.pdf. Retrieved 2026-10-06.
- ↑ 8.0 8.1 8.2 8.3 8.4 Douglas Lanman, Matthew Hirsch, Yun Hee Kim, Szymon Jakubczak, Ramesh Raskar (2015-12-17). "US20150362743A1 - Content-Adaptive Parallax Barriers for Automultiscopic Display". Google Patents. Massachusetts Institute of Technology. https://patents.google.com/patent/US20150362743A1/en. Retrieved 2026-10-06.
- ↑ 9.0 9.1 9.2 9.3 Mark Ellis Harris (2010-11-23). "3-D Without Four Eyes". IEEE Spectrum. https://spectrum.ieee.org/3d-without-four-eyes. Retrieved 2026-10-06.
- ↑ Frederic E. Ives (1902-01). "A novel stereogram". Journal of the Franklin Institute, vol. 153, no. 1, pp. 51-52. doi:10.1016/S0016-0032(02)90195-X. https://doi.org/10.1016/S0016-0032(02)90195-X. Retrieved 2026-10-06.
- ↑ 11.0 11.1 Clarence W. Kanolt (1918-03-26). "US1260682A - Photographic method and apparatus". Google Patents. United States Patent Office. https://patents.google.com/patent/US1260682A/en. Retrieved 2026-10-06.
- ↑ 12.0 12.1 12.2 Haruo Isono, Minoru Yasuda (1994-05-24). "US5315377A - Three-dimensional image display using electrically generated parallax barrier stripes". Google Patents. Nippon Hoso Kyokai (NHK). https://patents.google.com/patent/US5315377. Retrieved 2026-10-06.
- ↑ "2003 PC-RD3D 3D LCD Notebook PC". Sharp Corporation history. Sharp Corporation. https://global.sharp/corporate/info/his/only_one/i_c/subwin/i_c44.html. Retrieved 2026-10-06.
- ↑ 14.0 14.1 "Nintendo 3DS - Hardware Specs". Nintendo of America. Nintendo. 2011. https://web.archive.org/web/20111228135256/http://www.nintendo.com/3ds/hardware/specs. Retrieved 2026-10-06.
- ↑ 15.0 15.1 "HTC EVO 3D review: An extra dimension". GSMArena. 2011-08-19. https://www.gsmarena.com/htc_evo_3d-review-632p2.php. Retrieved 2026-10-06.
- ↑ "Acer Unveils SpatialLabs on ConceptD, Empowering Creators with Stereoscopic 3D". Acer Newsroom. Acer. 2021-05-27. https://news.acer.com/acer-unveils-spatiallabs-on-conceptd-empowering-creators-with-stereoscopic-3d. Retrieved 2026-10-06.
- ↑ Simon Baker (2025-04-22). "Samsung Odyssey 3D G90XF 4K 165Hz Monitor Officially Launched with Glasses-free 3D". TFTCentral. https://tftcentral.co.uk/news/samsung-odyssey-3d-g90xf-4k-165hz-monitor-officially-launched-with-glasses-free-3d. Retrieved 2026-10-06.
- ↑ Hideki Kakeya, Ayuki Hayashishita, Masaru Ominami (2018). "Autostereoscopic display based on time-multiplexed parallax barrier with adaptive time-division". Journal of the Society for Information Display, vol. 26, no. 10, pp. 595-601. doi:10.1002/jsid.717. https://doi.org/10.1002/jsid.717. Retrieved 2026-10-06.
- ↑ Robert Kooima, Andrew Prudhomme, Jurgen Schulze, Daniel Sandin, Thomas DeFanti (2010-11). "A Multi-viewer Tiled Autostereoscopic Virtual Reality Display". Proceedings of the ACM Symposium on Virtual Reality Software and Technology (VRST) 2010. Electronic Visualization Laboratory. doi:10.1145/1889863.1889899. https://www.evl.uic.edu/documents/vrst2010.pdf. Retrieved 2026-10-06.
- ↑ "AR Games: Augmented Reality". Nintendo UK. Nintendo. https://www.nintendo.com/en-gb/Hardware/Nintendo-3DS-Family/Instant-Software/AR-Games-Augmented-Reality/AR-Games-Augmented-Reality-115169.html. Retrieved 2026-10-06.
- ↑ Takuya Matsumoto, Kaoru Kusafuka, Goro Hamagishi, Hideya Takahashi (2018-05). "P-87: Glassless 3D Head Up Display using Parallax Barrier with Eye Tracking Image Processing". SID Symposium Digest of Technical Papers, vol. 49, no. 1, pp. 1511-1514. doi:10.1002/sdtp.12264. https://doi.org/10.1002/sdtp.12264. Retrieved 2026-10-06.
- ↑ 22.0 22.1 Douglas Lanman, Matthew Hirsch, Yunhee Kim, Ramesh Raskar (2010-12). "Content-adaptive parallax barriers: optimizing dual-layer 3D displays using low-rank light field factorization". ACM Transactions on Graphics, vol. 29, no. 6 (Proceedings of SIGGRAPH Asia 2010). doi:10.1145/1882261.1866164. https://doi.org/10.1145/1882261.1866164. Retrieved 2026-10-06.
- ↑ Gordon Wetzstein, Douglas Lanman, Matthew Hirsch, Ramesh Raskar (2012-07). "Tensor Displays: Compressive Light Field Synthesis using Multilayer Displays with Directional Backlighting". ACM Transactions on Graphics, vol. 31, no. 4 (Proceedings of SIGGRAPH 2012). doi:10.1145/2185520.2185576. https://doi.org/10.1145/2185520.2185576. Retrieved 2026-10-06.
- ↑ Fu-Chung Huang, Gordon Wetzstein, Brian A. Barsky, Ramesh Raskar. "Eyeglasses-free Display: Towards Correcting Visual Aberrations with Computational Light Field Displays". Stanford Computational Imaging Lab. doi:10.1145/2601097.2601122. https://www.computationalimaging.org/publications/eyeglasses-free-display-towards-correcting-visual-aberrations-with-computational-light-field-displays-siggraph-2014/. Retrieved 2026-10-06.